关键词: 低密度脂蛋白胆固醇, 非空腹血脂, Friedewald公式, Martine公式, Sampson公式, Vujovic公式, 心血管疾病

Abstract:

Objective To analyze the consistency of 4 calculated low-density lipoprotein cholesterol（LDL-C） levels（Friedewald formula，Martine formula，Sampson formula and Vujovic formula） with the direct determination of LDL-C（D-LDL-C） in non-fasting serum，and to evaluate the clinical value of LDL-C determination in non-fasting serum lipids. Methods Totally，110 cardiovascular disease（CVD） patients were enrolled，and the changes of fasting and postprandial 2 h serum lipids were compared. In addition，291 patients with CVD and/or diabetes mellitus（DM） and 259 healthy subjects were enrolled. Non-fasting serum lipids（fasting time <4 h） were determined，which included total cholesterol（TC），triglyceride（TG），LDL-C，high-density lipoprotein cholesterol（HDL-C） and small and dense low-density lipoprotein cholesterol（sd-LDL-C），and non-high-density lipoprotein cholesterol（non-HDL-C） was calculated. Spearman correlation analysis was used for D-LDL-C with calculated LDL-C levels，and the consistency was evaluated by different LDL-C cut-off points. Results For CVD patients，the levels of TC，HDL-C，LDL-C and non-HDL-C were lower after 2 h of routine meal（P<0.05），and there was no statistical significance in sd-LDL-C and TG levels whether postprandial or fasting condition（P>0.05）. For males，postprandial TG was increased as 0.11 mmol/L on average（P<0.05），and LDL-C was decreased as 0.13 mmol/L（P<0.05）. For females，the levels of TC，LDL-C and non-HDL-C were decreased as 0.22，0.14 and 0.19 mmol/L，respectively（P<0.05），and the change of TG level was not obvious（P>0.05）. Spearman correlation analysis showed that the calculated LDL-C levels of Friedewald formula，Martine formula，Sampson formula and Vujovic formula were positively correlated with D-LDL-C（r=0.922，0.953，0.949 and 0.944，P<0.05）. When TG≤2.26 mmol/L，the mean absolute deviation（MAD） between calculated LDL-C by Friedewald formula and D-LDL-C was minimum，and when TG>2.26 mmol/L，the deviation showed calculated LDL-C by Martine formula was minimum. For atherosclerotic cardiovascular disease（ASCVD） low-risk cases，different LDL-C cut-off points（2.6，3.4 and 4.9 mmol/L） were selected and concerned. When non-fasting serum TG>2.3 mmol/L，the consistency rates of calculated LDL-C of Martine formula，Sampson formula and Vujovic formula with D-LDL-C for evaluating ASCVD risk were 83.8%，83.8% and 82.5%，respectively，which were better than that of Friedewald formula（72.5%）. For ASCVD high-risk cases，when TG<2.3 mmol/L，the clinical assessment results of 4 calculated LDL-C levels showed good consistency with D-LDL-C（≥95%）. When 2.3<TG<5.6 mmol/L，the calculated LDL-C level of Martine formula showed the highest consistency with D-LDL-C（88.8%）. When TG>5.6 mmol/L，all the calculated LDL-C levels had no consistency with D-LDL-C. Conclusions Compared with fasting serum lipids，the levels of non-fasting serum lipids are usually increased in TG level and decreased in TC and LDL-C levels，but there was difference for males and females between fasting and postprandial changes. Using non-fasting calculated LDL-C of Martine formula may do help for the precise estimation of LDL-C level by clinicians，especially when TG>4.52 mmol/L. Non-fasting condition for lipid determination is convenient and acceptable for hypertriglyceridemia and CVD and DM patients，which should choose fasting blood sampling for serum lipid determination.

Key words: Low-density lipoprotein cholesterol, Non-fasting lipid, Friedewald formula, Martine formula, Sampson formula, Vujovic formula, Cardiovascular disease